Aligning AI
Central Problems in AI Safety
23
Learning Values
24
Power-Seeking
25
Agent Foundations
26
Learning from Humans
27
Decomposing Tasks
28
Interpretability
29
Governance